notice
This is documentation for Rasa X Documentation v0.35.x, which is no longer actively maintained.
For up-to-date documentation, see the latest version (1.1.x).
Quick-Installation
The simplest and quickest way to install Rasa X on a server or cluster is to use our quick-install script.
To install Rasa X, simply run the command below in a compatible environment (see Requirements):
When done, the script will print the URL to access Rasa X:
The script will perform the following steps:
Install an embedded Kubernetes cluster. If the Kubernetes command-line interface
kubectl
is already configured, it will use the configured cluster. Please see Existing Cluster Installation for more details if you are interested in using the script with an existing Kubernetes cluster.Install the Helm command-line interface.
Install Rasa X using the Rasa X Helm Chart.
Requirements
Hardware & OS Requirements
note
Rasa X is intended to be deployed on a server and not to a personal/local machine. Deploying on a server is recommended because Rasa X is designed to stay up continuously, and not to be frequently stopped or restarted.
The installation script is officially supported on the following operating systems:
Ubuntu 16.04 / 18.04
Debian 9 / 10
Red Hat 7 / 8
CentOS 7 / 8
You can also use this script to run Rasa X on Windows or MacOS using Multipass.
For any other Linux operating systems, please follow the Docker Compose Manual Installation instructions.
Embedded Cluster Requirements
In case you don’t have an existing Kubernetes Cluster, the installation script will set up an embedded Kubernetes cluster on your machine. To run Rasa X on your machine, it should have the following hardware specs:
vCPUs
Minimum: 2 vCPUs
Recommended: 2-6 vCPUs
RAM
Minimum: 4 GB RAM
Recommended: 8 GB RAM
Disk Space
- Recommended: 50 GB disk space available
Supported Browsers
The web interface aims to support browsers that meet the following criteria:
0.2% market share
not Internet Explorer
not Opera Mini
Installation
Embedded Cluster Installation
Before installing using the script you’ll need a deployment environment that can run Rasa X. Please see the Requirements for the requirements.
Once you are ready, run the install command:
When done, the script will print the URL to access Rasa X:
Rasa X will be installed in the Kubernetes namespace rasa
, with a Helm release name rasa
.
Existing Cluster Installation
You can also use this installation method with an existing Kubernetes cluster.
The existing cluster has to configured with the Kubernetes command-line-interface
kubectl
. To see if kubectl
is installed on your machine and if it is correctly
connected to a cluster run the following command:
If you get an error kubectl: command not found
or there is no output for the
Server Version
either connect to a Kubernetes cluster or use the
Embedded Cluster Requirements.
Once you are correctly connected to the cluster, simply run the installation command:
Windows & MacOS Installation
The recommended way of installing Rasa X on Windows or MacOS using the quick-install script is with Multipass. Multipass is a mini-cloud that provides a command line interface to launch and manage Ubuntu instances.
First of all, you have to download Multipass from the official website and install it. Alternatively, you can use your package manager like brew or chocolatey.
After you installed it, you can create and launch an Ubuntu instance and access it using the following commands:
Now you can use the quick-install script to install Rasa X:
After you installed it, wait for Kubernetes to spin on all the instances and execute the following on your local machine to get the IP address of the running Multipass Ubuntu instance:
Use the IP4 address you’ve got to access Rasa X in your browser.
Go to Next StepsAccessing Secrets
To access auto-generated passwords later on, head over to the cluster deployment
docs on Accessing Secrets. Both the release_name
and
namespace
is rasa
.
Accessing Logs
To access logs of the pods, head over to the cluster deployment
docs on Accessing Logs. Both the release_name
and
namespace
is rasa
.
Uninstalling
To uninstall Rasa X, run the script with the --uninstall
flag.
Next Steps
Connect a custom action server if you are using custom actions.
Set up Integrated Version Control to connect your Rasa X instance to a remote Git repository.
Deploy your assistant using Rasa X.